作 者 ：孟丹,李小娟*,宫辉力,赵文吉
期 刊 ：生态学报 2010年 30卷 13期 页码：3491~3500
关键词：热力景观格局；城市热环境效应；公园景观；道路廊道；北京；
Keywords:thermal environment landscape pattern, urban thermal environment effect, park landscape, road corridor, Beijing,
摘 要 ：城市热环境是城市生态环境中的一个重要指标，将景观生态学理论融入到热环境研究中，尝试探讨北京地区热力景观格局及城市公园、道路景观的热环境效应。地表温度反演是分析热力景观格局及典型城市景观热环境效应的前提，论文以北京地区为例，首先利用两景ASTER影像数据采用TES算法定量反演地表温度。通过半变异函数分析地表温度空间异质性，确定最大采样尺度，然后在景观统计软件Fragstats中，计算不同粒度下的景观格局指数，分析热力景观格局及其尺度效应。通过景观斑块特征分析和缓冲区分析，探讨公园景观斑块、道路景观廊道特征的热环境效应。总体上公园景观对应的平均温度随着公园面积、边界长度的增加而减小，随着公园周长面积比增大而增大；随着距离公园渐远，地表温度升高，且升温趋势变缓。随着道路密度增加，道路平均温度显著升高，标准差显著降低，道路密度等级与道路平均温度的相关系数达到0.8021；随着距离道路中心线距离增加，缓冲区内的平均温度略有下降，但变化微弱。因此，应充分重视公园景观在缓解城市热环境方面的作用，合理布局城市道路。
Abstract:Study on thermal environment and thermal effect is one of the most important topics in city climate and environment researches. The paper integrates some theories of ecology landscape, geostatistics technology and spatial analysis technology of GIS into the research on thermal environment issues, and explores the thermal landscape pattern, the thermal effect of parks landscape and road corridors in Beijing. As retrieving land surface temperature (LST) and analysis of its distribution is of great efficiency to research thermal environmental problems of earth surface, two ASTER imageries are used for the LST retrieval by TES algorithm. Fitting of the whole day′s continuous ground temperature with meteorological observations data at the same date when the satellite transit using sinusoidal function is utilized, and the following interpolation generates ground temperature at satellite transit time to verify the retrieved LST. The overall temperature accuracy is less than 0.5℃. Basing on the LST variable, the thermal landscape pattern is studied by calculating landscape pattern indices under different grid size in Fragstats software. Considering the scale effect of thermal landscape heterogeneous pattern, the largest sampling scales are determined by the semi-variant function. The results show that the largest sampling scale is 818.392m and 778.520m in the urban scene and the suburb scene, respectively. Suburban thermal landscape patches are continuous, while the urban thermal landscape patches are relatively dispersed. This illustrates that thermal landscape has a close relationship with built-up levels of urban development and construction. With the increase of sampling particle size, the patch density and contagion index show a downward trend, and the average patch area increases exponentially, while the diversity index changes little. Except the diversity index, the other three indices vary with different grain size. In addition, the great importance of choosing appropriate resolution should be noted for the thermal landscape studies. Studies have shown that large area of vegetation cover and water can relieve and regulate the thermal environment. What′s more, the urban parks is the assemblage of green space and water bodies. Therefore, park is undoubtedly the city′s ‘cold island’. In contrast, due to the physical properties and heat emissions from vehicles, road corridor is the relative heat strip′. For studying thermal environment effect which is of great significance for urban planning, park and road landscape are chosen for the study of thermal environment effect, by constructing the models between landscapes patches characteristics and the corresponding average LST, and by calculating the LST varies in the buffer zones of landscapes themselves. The thermal environment effect analysis draws the conclusion that generally the average LST decreases with the increase of area, perimeter of park, and the decrease of ratio of perimeter to area. In areas further away from the parks, the LST raised, and the raising trend slowed down. Specifically, the average LST increase is 0.5366℃ from the 0-90m buffer zone to 90-180m buffer zone outside the park, while the one is 0.3258℃ from the 90-180m buffer zone to the 180-270m buffer zone. As road density increases, the average of LST increases and standard deviation of LST decreases significantly. The correlation coefficient between the road density ranks and LST of roads was 0.8021. The average LST of the buffer zone decreased slightly with distance from the road centerline. Therefore, more attention is ought to paid to the rational distribution of urban roads and the landscape park for the sake of easing the implication of urban heat environment pressure.
全 文 ：